SLAM算法与工程实践系列文章

下面是SLAM算法与工程实践系列文章的总链接，本人发表这个系列的文章链接均收录于此

SLAM算法与工程实践系列文章链接

下面是专栏地址：

SLAM算法与工程实践系列专栏

文章目录

  • SLAM算法与工程实践系列文章
    • SLAM算法与工程实践系列文章链接
    • SLAM算法与工程实践系列专栏
  • 前言
  • SLAM算法与工程实践——相机篇：传统相机使用（1）
    • 相机相关命令
    • 出现的问题
    • 调用相机
      • 以MJPG格式打开相机

前言

这个系列的文章是分享SLAM相关技术算法的学习和工程实践

SLAM算法与工程实践——相机篇：传统相机使用（1）

相机相关命令

插上USB相机，使用命令查看USB设备

lsusb

可以识别相机

使用命令查看识别到几个摄像头

ll /dev/video*

然后改变其权限

sudo chmod 777 /dev/video0sudo chmod 777 /dev/video1

安装 v4l-utils 工具包

sudo apt-get install v4l-utils

使用下列命令查看相机参数

v4l2-ctl -d /dev/video0 --list-formats-ext

在命令行输出为：

ioctl: VIDIOC_ENUM_FMTType: Video Capture[0]: 'MJPG' (Motion-JPEG, compressed)Size: Discrete 2560x800Interval: Discrete 0.017s (60.000 fps)Interval: Discrete 0.033s (30.000 fps)Interval: Discrete 0.067s (15.000 fps)Size: Discrete 2560x720Interval: Discrete 0.017s (60.000 fps)Interval: Discrete 0.033s (30.000 fps)Size: Discrete 1280x800Interval: Discrete 0.008s (120.000 fps)Interval: Discrete 0.017s (60.000 fps)Interval: Discrete 0.033s (30.000 fps)Size: Discrete 1280x792Interval: Discrete 0.008s (120.000 fps)Interval: Discrete 0.017s (60.000 fps)Interval: Discrete 0.033s (30.000 fps)Size: Discrete 1280x400Interval: Discrete 0.008s (120.000 fps)Interval: Discrete 0.017s (60.000 fps)Interval: Discrete 0.033s (30.000 fps)Size: Discrete 640x400Interval: Discrete 0.005s (210.000 fps)Interval: Discrete 0.008s (120.000 fps)Interval: Discrete 0.017s (60.000 fps)Size: Discrete 640x392Interval: Discrete 0.005s (210.000 fps)Interval: Discrete 0.008s (120.000 fps)Interval: Discrete 0.017s (60.000 fps)[1]: 'YUYV' (YUYV 4:2:2)Size: Discrete 2560x800Interval: Discrete 0.500s (2.000 fps)Size: Discrete 2560x720Interval: Discrete 0.500s (2.000 fps)Size: Discrete 1280x800Interval: Discrete 0.200s (5.000 fps)Size: Discrete 1280x792Interval: Discrete 0.200s (5.000 fps)Size: Discrete 1280x400Interval: Discrete 0.067s (15.000 fps)Interval: Discrete 0.100s (10.000 fps)Size: Discrete 640x400Interval: Discrete 0.033s (30.000 fps)Size: Discrete 640x392Interval: Discrete 0.033s (30.000 fps)

可以看出本相机的参数和对应设置下的帧率，最高为 120 帧，分辨率为 2560×800，我这选用 1280×400

接下来看看你的USB摄像头支持什么附加属性，只需要执行下面的命令即可

v4l2-ctl -d /dev/video0 --list-ctrls

得到下面的结果

User Controls brightness 0x00980900 (int): min=-64 max=64 step=1 default=0 value=0 contrast 0x00980901 (int): min=0 max=95 step=1 default=0 value=0 saturation 0x00980902 (int): min=0 max=100 step=1 default=64 value=64hue 0x00980903 (int): min=-2000 max=2000 step=1 default=0 value=0white_balance_automatic 0x0098090c (bool) : default=1 value=1gamma 0x00980910 (int): min=100 max=300 step=1 default=100 value=100

使用应用程序茄子（cheese）

输入命令：

sudo apt-get install cheese

装好后，用命令：cheese，即可打开。如果指定打开video1，输入命令：

cheese -d /dev/video1

读取相机配置文件 HBV-1780-2.yaml

int UseStereoCam::loadCamParam(std::string cfgPath) {cv::FileStorage fs(cfgPath, cv::FileStorage::READ);if (fs.isOpened()) {std::cout << "Stereo Camera Paramter is loading..." <> IntrinsicMatrixLeft;//存放内参矩阵,左相机IntrinsicMatrixLeft.at(0, 0) = IntrinsicMatrixLeft.at(0, 0) * resizeScalse;IntrinsicMatrixLeft.at(1, 1) = IntrinsicMatrixLeft.at(1, 1) * resizeScalse;IntrinsicMatrixLeft.at(0, 2) = IntrinsicMatrixLeft.at(0, 2) * resizeScalse;IntrinsicMatrixLeft.at(1, 2) = IntrinsicMatrixLeft.at(1, 2) * resizeScalse;fs["IntrinsicMatrixRight"] >> IntrinsicMatrixRight;//存放内参矩阵,右相机IntrinsicMatrixRight.at(0, 0) = IntrinsicMatrixRight.at(0, 0) * resizeScalse;IntrinsicMatrixRight.at(1, 1) = IntrinsicMatrixRight.at(1, 1) * resizeScalse;IntrinsicMatrixRight.at(0, 2) = IntrinsicMatrixRight.at(0, 2) * resizeScalse;IntrinsicMatrixRight.at(1, 2) = IntrinsicMatrixRight.at(1, 2) * resizeScalse;//畸变系数fs["DistCoeffLeft"] >> DistCoeffLeft;//畸变矩阵，左相机fs["DistCoeffRight"] >> DistCoeffRight; //畸变矩阵，右相机//旋转矩阵 平移向量fs["RotationMatrix"] >> R;fs["TranslationVector"] >> T;T = T * resizeScalse;fs.release();leftFocalLength =(IntrinsicMatrixLeft.at(0, 0) + IntrinsicMatrixLeft.at(1, 1)) / 2.0;rightFocalLength =(IntrinsicMatrixRight.at(0, 0) + IntrinsicMatrixRight.at(1, 1)) / 2.0;// 处理焦距（例如取平均值）focalLength = (leftFocalLength + rightFocalLength) / 2.0;fx = (IntrinsicMatrixLeft.at(0, 0) + IntrinsicMatrixRight.at(0, 0)) / 2.0;fy = (IntrinsicMatrixLeft.at(1, 1) + IntrinsicMatrixRight.at(1, 1)) / 2.0;cx = (IntrinsicMatrixLeft.at(0, 2) + IntrinsicMatrixRight.at(0, 2)) / 2.0;cy = (IntrinsicMatrixLeft.at(1, 2) + IntrinsicMatrixRight.at(1, 2)) / 2.0;baseline = cv::norm(T) / 100; //基线（baseline）单位设置为分米std::cout << "camera parameter loaded successfully!" << std::endl;return 0;} else {std::cout << "Failed to load camera parameter!" << std::endl;return -1;}}

出现的问题

参考：

解决OpenCV的GStreamer warning警告

在用opencv调用摄像头时报错

[ WARN:0] global ../modules/videoio/src/cap_gstreamer.cpp (935) open OpenCV | GStreamer warning: Cannot query video position: status=0, value=-1, duration=-1[ WARN:0] global ../modules/videoio/src/cap_gstreamer.cpp (1758) handleMessage OpenCV | GStreamer warning: Embedded video playback halted; module v4l2src0 reported: Device '/dev/video0' failed during initialization[ WARN:0] global ../modules/videoio/src/cap_gstreamer.cpp (515) startPipeline OpenCV | GStreamer warning: unable to start pipeline[ WARN:0] global ../modules/videoio/src/cap_gstreamer.cpp (1057) setProperty OpenCV | GStreamer warning: no pipeline[ WARN:0] global ../modules/videoio/src/cap_gstreamer.cpp (480) isPipelinePlaying OpenCV | GStreamer warning: GStreamer: pipeline have not been created摄像头打开失败!

解决办法：

修改打开摄像头的代码如下：

cap.open(0,CAP_V4L2);

调用相机

在编写多个主函数共用src和include定义的类的文件时，CMakeLists.txt的编写应该类似下面的写法，然后包含每个子项目的子文件夹

cmake_minimum_required(VERSION 2.8)project(DEPTH_COMPUTE)IF (NOT CMAKE_BUILD_TYPE)SET(CMAKE_BUILD_TYPE Release)ENDIF ()MESSAGE("Build type: " ${CMAKE_BUILD_TYPE})# set pathset(LIB_PATH ${PROJECT_SOURCE_DIR}/lib)set(HEAD_PATH ${PROJECT_SOURCE_DIR}/include)set(EXEC_PATH ${PROJECT_SOURCE_DIR}/bin)set(DEPTH_LIB calc_depth) # set libiary namefind_package(OpenCV 4 REQUIRED)find_package(Eigen3 3.1.0 NO_MODULE)find_package(Pangolin REQUIRED)include_directories(${HEAD_PATH})aux_source_directory(${PROJECT_SOURCE_DIR}/src LIB_SRC)SET(LIBRARY_OUTPUT_PATH ${LIB_PATH})add_library(${DEPTH_LIB} SHARED ${LIB_SRC})TARGET_LINK_LIBRARIES(${DEPTH_LIB}${OpenCV_LIBS}${EIGEN3_LIBS}${Pangolin_LIBRARIES})# add subdirectories 这里是具体的主函数add_subdirectory(useCamForShot) add_subdirectory(openCamera)

子项目的CMakeLists.txt应该如下

cmake_minimum_required(VERSION 2.8)project(OPENCAM)find_package(OpenCV 4 REQUIRED)include_directories(${HEAD_PATH})link_directories(${LIB_PATH})set(CMAKE_BUILD_TYPE "Release")set(CMAKE_CXX_FLAGS "-O3")set(EXECUTABLE_OUTPUT_PATH${EXEC_PATH})#add_library(${DEPTH_LIB} SHARED${SRC_LIST})add_executable(open_camera open_camera.cpp)target_link_libraries(open_camera ${DEPTH_LIB} ${OpenCV_LIBS})

以MJPG格式打开相机

参考：

OpenCV+V4L实现MJPG格式拉取USB摄像头

关于openCV VideoCapture获取摄像头MJPG视频的问题

相机支持的视频流格式

在打开相机后，设置参数

capture.open(camIndex, cv::CAP_V4L);// open camera//设置编码格式，这里只能写在分辨率和帧率后面，否则无效capture.set(cv::CAP_PROP_FOURCC, cv::VideoWriter::fourcc('M', 'J', 'P', 'G'));//视频流格式